I I 111

1. 13. 2.50. [CONTRIBUTION FROM THE CHEMIW RESEARCH LABORATORY. OF POLAROID CORPORATION]. The Absorption Spectra of Certain Aldazines...
17 downloads 0 Views 474KB Size
THEABSORPTION SPECTRAOF CBRTAIN ALDAZINES

Jan., 1945

[CONTRIBUTION FROM THE CHEMIW

13

RESEARCH LABORATORY OF POLAROID CORPORATION]

The Absorption Spectra of Certain Aldazines BY ELW R. BLOUTAND RALPHM. GOFSTEIN The ultraviolet spectra of the azines obtained from benzaldehyde, cinnamaldehyde and 5phenylpentadienal have been recorded by previous workers,' who have noted the effect of the nitrogen atoms in removing the fine structures of the bands associated with the corresponding diphenylpolyenes. The absence of fine structure simplifies the study of the effect of substituent groups upon spectra. In the present communication, the ultraviolet absorption spectra of substituted aldazi'nes have been examined with a view toward extending our knowledge of the correlation of light abdorption and structure. We are concerned with the spectra of compounds of the general type

-

-

-

1

/-*\

R-(CH=CH)=-CH=N-N=CH-(CH=CH)=-R

where R represents an aromatic group, and x is either 0 or 1. Since many substituted aromatic aldehydes are available, aldazines of the above type may be prepared conveniently by reaction with hydrazine. We have reeamined the spectra of benzalazine and cinnamalazine (Fig. 1) and have found the maxima to be at 300 and 352 mp in agreement with Radulescu and Alexa.'12 A comparison of the spectra of 1,l'-naphtfialazine with benzalazine shows the expected displacement of the absorption bands toward longer wave lengths characteristic of a more extensively conjugated resonating system, The azine from furfuraldehyde shows an absorption maximum a t 335 mp, which is a displacement of 35 mp as compared with benzalazine. This may be explained by the relatively large contribution of dipolar structures such as I and 11.

2.50

\

1

I

I

I

1

1

I

I

I

I

I

I

I

1

1

1

I

1

1

I

I

I

I

I

I

'% 1

1

I

I

. I

i

I d

-

f L

+ The spectra of the isomeric nitrobenzalazines (Fig. 2) indicate the effect of positional substitution on light absorption. The p-substituted compound) having a maximum a t 318 mp, shows a (1) Radulescu and Alexa, B e y . , 64, 2230 (1931). (2) The extinction coefficients, however, while appearing to differ in the two determinations, are actually similar, the difference being in the definition of e.

3.00 2.50

I

1

260

I

I

I

I

111

300 350 400 X in mp. Fig. 2.,4,4'-dinitrobenzalazine; -.-. -.-, 3,3'-dinitrobenzalazine; - - - - - - - - - ,2,2'-dinitrohenzalazine.

ELKAN R.BLOUTAND RALPHM. GOFSTEIN

14

bathochromic displacement compared to benzalazine. 3,3’-Dinitrobenzalazine shows no welldefined maximum but rather a rudimentary separation at the band head from 265 to 295 mp. The o-nitro compound, 2,2’-dinitrobenzalazine, shows two distinct absorption bands; the shorter wave length band exhibits fine structure, whereas the band at longer wave length shows a single maximum at 332 mp. Analogous phenomena are observed with the isomeric hydroxybenzalazines (Fig. 3). The phydroxy compound has a single absorption band with a maximum at 335 mp; whereas the mhydroxyazine has a maximum at 300 mp and an inflection point at 925 mp. In the o-hydroxy compound, on the other hand, there is complete separation into two bands, the maxima lying at 295 mp and 355 mp, respectively.

-I

\

t

Vol. 67

.

/

W

*.

i

1

Pig.4.-. . . . . . . . ., 1,l’-naphthalazine;4,4’dihydroxy-1,l’-naphthalazine;- - - - - - - - -, 2,2’dihydroxy-1,l’-naphthalazine.

The recent correlation^^^^ between resonance and color in organic substances afford an explanation of these results. Color may be attributed to transitions to excited sthtes of high dipole moment. Therefore, any substituents affecting the polarizability (or the dipole moment of the excited states) will affect the light absorption. Furthermore, planar molecules may be considered to have two polarizabilities at right angles to each other, and compounds such as benzalazine may be treated as linear when one considers the limit structures of the excited state, 111.

i i

‘i

3.001 2.50

I

I I l m

250

/

/

/

~

300

~

~

350

~

~

1



1



1

~

400

A in mp. Fig.3.,4,4‘-dihydroxybenzalazine;-.-.-.-, 3,3’-dihydroxybaizaIazi1le;- - - - - - - - - , 2,2’dihydroxybenzalazine.

_ .

Likewise, the symmetrically p-substituted benzalazines may be considered as linear since the effect of the substituents will be directed along the same This same effect is apparent upon consideration axis. Therefore, only positional differences in of the spectra of 4,4‘-dihydroxy-l, 1’-naphthal- absorption spectra would be expected between unazine and 2,2‘-dihydroxy-1,l’-naphthalazine (Fig. substituted benzalazine and the p-substituted benzalazines. This is, in fact, the case (cf. Figs. 4). The unsubstituted naphthalazine shows a 1,2,3,5). This line of reasoning may be extended single peak a t 355 mp, but in the 4-hydroxy com- to explain the differences in the position of the pound the maximum ’is displaced to 385 m p and maxima of the p-substituted benzalazines. As,nows no inflection points or band splitting. The suming transitions to forms such as I11 are respondisplacement is of about the same magnitude observed with the corresponding benzal compounds sible for the absorption maximum a t 300 mp of (cf. Figs. 1 and 3). 2,2’-Dihydroxy-l1l’-naph-benzalazine, then forms such as IV (3) Lewis and Calvin, Chem. Rev.. %6, 273 (1939); Bury, Tms thalazine, on the other hand, shows two absorp57, 2115 (1935); Pauling, Pmc. Nall. Accrd. Sci., 16, 577 tion maxima, one at 330 mp and the other at 409 JOURNAL, (1939); Mulliken, J . Chcm. Phys., 7 , 121,364, 570 (1939). mp, the same region in which the corresponding (4) Cf.Pauling in “Organic Chemistry,” H. Gilman, Editor, 1943, 4-hydroxy compound shows only one. 2nd ed., Vol. 11, pp. 1944 ff.

THEABSORPTION SPECTRA OF CERTAIN ALDAZINES

Jan., 1945

IV \ C H ~ - O H

must account for the maximum at 335 mp of the p-hydroxy compound. In effect the hydroxyl groups increase the length of the absorbing system and thus shift the maximum to longer wave lengths.6 In the corresponding #-nitro compound, the single absorption peak may be attributed to the preponderance of essentially linear forms such as V, VI, VI1 and VIII.* O\k

e o/

\ C H G G / O '

-\N=N

-

V O ' \&-=CH

o//

-

\O

oe

\N=N

\cH=-&/

-

VI

'00

e o Y = O = C H

€30

-

b

VI1

eo ) k - a = C H

€30

\N=$=cH-

15

mp involves more nearly linear forms such as X in which the nitro groups add to the electron mobility of the system. m-Nitrobenzalazine exhibits incomplete band separation. The maximum at 295 mp is again attributed to linear structures analogous to the unsubstituted compound (cf. III), whereas the lower wave length maximum probably involves forms similar to IX, the band occurs in the same region as the maximum observed in nitr0benzene.l Spectra of the same type as those of the nitro compounds are observed in the 0- and m-hydroxybenzalazines, but the maxima lie a t much longer wave lengths. For example, 2,2'-dihydroxybenzalazine has absorption bands at 295 mp and 355 mp. It is apparent that the shorter wave length band is due to resonance forms similar to those responsible for the absorption of benzalazine (111), while the longer wave length band may be attributed to non-colinear forms. The difference between the effect of the nitro and hydroxyl groups may be attributed to their electronk properties. The hydroxyl group tends to give up electrons to the aromatic ring, whereas the nitro group tends to draw electrons from the ring.* Substituents such as hydroxyl or methoxyl that tend to add electrons will have a bathochromic effect on benzalazine while groups that take out electrons will have a hypsochromic effect. It should be pointed out that double maxima have been noted in the spectra of o-nitroaniline, onitrophenol, o-nitrotoluenes and o-hydroxy- and o-methoxybenzaldehyde.'O The possibility of other explanations for the band splitting in the o-substituted ComDounds has bein conGdered since it is evident thai chelation may occur in the o-hydroxyazines. Other worker^^^^^ have attributed double maxima observed in absorption spectra to inter- and intramolecular hydrogen bonding. That chelation of the o-hydroxyl group and the azine nitrogen in 2,2'-dihydroxybenzalazine is not the cause of the band splitting observed is shown by examination of the spectrum of the corresponding methyl ether,

a-N\, +No

/

VI11 In the o-substituted compounds, resonance structures are permitted which involve charge oscillation non-colinear with the main chromophoric axis and forms such as I X for o-nitrobenzalazine may contribute to an excited state. I

T

It is not impossible that the peak at 255 mp' is due to transitions to excited states corresponding to forms such as IX, whereas the maximum a t 325

(8) I t is interesting to note that one of the main resonance forms of 2,2'-dihydroxybenzalazinecorresponds closely in electronic configuration to that of o-nitrophenol; that is L

/H

(5) C f . Hausser, Kuhn and Seitz, 2. physik. Chcm., B49, 391 (1935). (6) I t is recognized that, since 4,4'-dinitrobenzalazine does not absorb at as high wave lengths as the corresponding hydroxy azine, the importance of forms that contribute to the lengthening of the conjugate absorption must be less, e. g., forms VI and VII. T h e smaller contribution and instability of such forms are in turn probably due to the presence of adjacent positive charges. (7) Cf.Absorption spectrum of nitrobenzene; maximum 255 mp; Wolf and Herold, 2. physik. Chcm., B18,201 (1931).

+P@

where the C-N is replaced by the N=O. o-Nitrophenol hM an absorption maximum at 355 mps as does 2,2'-dihydrorybenzalazine. We attribute this maximum to transitions to forms such as that above which are obviously non-colinear with resonance forms such as

to which we attribute the maximum at 295 mp. (9) Dede and Rosenberg, Ber., 67, 147 (1934). (10) Morton and Stubbs, J . Chem. Soc., 1347 (1940). (11) Shepherd and Newsome, THISJOURNAL, 64, 2937 (1942).

ELKANR.BLOUTAND RALPHM. GOFSTEIN

16

Vol. 67

TABLE I M. p. , O C . reported

Observed rn. p . , OC. (cor.)

Benzalazine Cinnamalazine Furfuralazine 1,1‘-Naphthalazine 4,4’-Dinitrobenzalazine 3,3‘- Dinitrobenzalazine 2,2’-Dinitrobenzalazine

93a 162“ 111-1 12h 156“ 297d 196-1 9rd 205d

92.5-93 170- 171 110-111.6 156-157 297-298 196-197 ao5-a06

Pyridine Ethanol Ethanol Ethanol Pyridine Butanol Dioxane

4,4’-Dihydroxybenzalazine 3,3’-Dihydroxybenzakazine

c. 268“ 204-205’

270 (dec.) 204-205

Ethanol Ethanol

2,2’-Dihydroxybenzalazine

214’

2 13-2 14

Ethanol

4,4’-Dihydroxy-l, 1’-naphthalazine

236A

239-240 (dec.)

Dioxane

2,2 ’-Dihydroxy- 1,l ’-naphthalazine

>290i

>300

Butanol

2,2’-Dimethoxybenzalazine

143O

143-144

Ethanol

Compound

4,4 ’-Dimethoxybenzalazine

168-177d 170- 185 liquid crystalline liquid crystalline

-

Cr stn soLen;

Pyridine

L

x

.

in rnM

300 352 335 354 318 265-295 260 332 336 300 ~ 3 2 5 295 355 245 385 232 267 330 409 293 340 330

log

c

4.56 4.76 4.60 4.48 4.53 4.51-4.53 4.32 3.98 4.63 4.46 4.40 4.40 4.36 4.33 4.29 4.77 4.43 4.16 4.51 4.11 4.29 4.56

indicates an inflection point. Curtius and ay, 1.prakt. Chem., [2] 39, 27 (1889). * Minunni, Gazz. chim. ital., 29, 11, 467 (1899). c Curtius, J. prakt. Chem., [ Z j S S , 393 (1912) Knopfer, Motaatsh., 30, 29 (1909). Vorlander, Bel., 39, 803 (1906). f Noelting, Ann. chim. phys., [8]19,497 (1910). 0 Borsche, Ber., 34, 4297 (1901). a Gattermann, Ann., 357, 313 (1907). * Gattermann and von Horlacher, Ber., 32, 284 (1899). a

with the o-hydroxyazine (Fig. 3). We have also determined the spectrum of 4,4’-dimethoxybenzalazine and observed that there is only a single peak (Fig. 5). This curve again closely resembles the corresponding hydroxy compound except for a displacement of the maximum about 5 mp to shorter wave lengths, which indicates a smaller contribution of the methoxyl group to dipolar structures, and hence its absorption band lies farther out in the ~ltraviolet.~ Our viewpoint that the double peak observed in the azines substituted in the 2,2‘-positions is not due to hydrogen bonding is further substantiated by the incipient band splitting shown in the spectra of the m-substituted benzalazines (Figs. 2 and 3). Intramolecular association in these compounds is impossible on steric grounds,12 and, therefore, it may be assumed that the greater approach toward linearity of the m-substituted azines as compared with the 0 - accounts for the incomplete separation of the absorption bands in I the former compounds. \i possibility that the single peaks observed ‘1 i in The unsubstituted and $-substituted azines are I i in the reality composed of two tands at the same ‘\ - 1 wave length has been consi. ered. That this is \ I 1 probably not the case is evident from a considera‘ 1 tion of the spectra of 2,2f-dihydroxy-l,lf-naphthalazine a.id 4,4‘-dihydroxy- 1,1 -naphthalaziqe (Fig. 4). It is found that one of the maxima of

2,2’-dimethoxybenzalazine (Fig. 5 ) , in which compound chelation does not exist. The two maxima, a t 290 mp and 340 mp, are displaced slightly toward shorter wave lengths as compared

t 4.50

4.00 M

s 3.50 3.00 -

-

2.50

I

I

I

I

I

I

I

1

1

1

1

1

1

1

1

(12) Cf.Pauling, “Nature of the Chemical Bond,” Chap. IX, Corne11 Univ. Press, Ithaca, N. y . , 2nd edition, 1942.

Tan., 1945

THEEFFECTSOF TEMPERATURE ON THE

the 2-hydroxylates compound has a higher extinction coefficient (log e, 4.51) than the single peak of the 4-hydroxylated compound (log e, 4.29) , which in turn is lower than the corresponding naphthalazine containing no hydroxyl groups (log e, 4.48). If the curves that show only a single maximum were composed of two bands lying at the same place, it would be expected that the single peak would have a higher extinction coefficient than those of the peaks found in double peaked curves. Since one of the peaks of the doubly peaked curve is higher, we may assume that we are dealing with single absorption bands in the unsubstituted and p-substituted benzalazines.la (13) We have not neglected t o consider that the shorter wave length band in the spectra showing two maxima might be a second order band. 4,4‘-Dihydroxybenzalazineis more closely a linear harmonic oscillator than the isomeric 3-hydroxy or 2-hydroxy compound on purely geometric grounds. Thus,assuming t h a t the single absorption band of the 4-hydroxy compound is due t o a transition u 0 + u = 1 ( i , e., a transition from the ground state t o the first excited state) it is possible t h a t the double maximum observed in the o isomer is actually the result of two transitions, u = 0 --C v 1 and v 0 -* Y 2 which are due t o oscillations along the same direction in the molecule. The probability of a second order transition occurring in the $-hydroxy compound is zero; Lewis and Bigeleisen, THIS JOURNAL, 60,2107 (1943). The o-isomer has more or less anharmonicity due perhaps t o resonance contributions involving charge oscillation non-colinear with the main chromophoric axis. This anharmonicity would make the probability of the transition w 0 --C Y 2 relatively greater since the electronic levels would lie lower and closer together. An analogous argument would hold for the m: compounds with consideration of the fact that the mesomeric effect of the azimethylene chain would not give rise to so great an interaction as t h a t expected for the o- and p-compounds so that the splitting would be less pronounced. Also the m-compound would presumably

-

-

-

-

[CONTRIBUTION FROM

THE

-

POLYMERIZATION OF STYRENE

17

Acknowledgment.-The authors wish to express their appreciation to Dr. R. B. Woodward for many helpful discussions and to Mrs. D. C. Silverman, Mr. A. S. Makas, and Mr. E. Farmer for their valuable assistance in the determination of most of the absorption spectra recorded herein. Experimental The azines were prepared by shaking 2.3 moles of the corresponding aldehyde with 1 mole of 85% hydrazine hydrate in water or dilute ethyl alcohol. In some cases the hydrazone rather than the azine separated, but on addition of a drop of dilute hydrochloric acid to the solvent used in the recrystallization, the hydrazone was decom posed to the azine.

The absorption spectra measurements wert made on a Beckmann quartz spectrophotometer model DU using a 1-cm. quartz cell and a hydrogen discharge tube as an ultraviolet source. Absolute ethyl alcohol was used as a solvent throughout. s=arY The ultraviolet absorption spectra of several substituted aromatic aldazines have been determined. The band splitting noted in 2,2‘-substituted benzal- and naphthalazines has been attributed to new absorption bands arising from the non-linearity of such molecules as compared with the unsubstituted and p-substituted azines. be a more nearly linear harmonic oscillator than the o-compound. But the extinction coefficients for the two bands are nearly equal, and the ratio of their frequendies is